The total memory size is hardcoded again because the protocol does not
allow to retrieve the correct memory layout in advance. As long as there
is no device with a different amount of memory, that should be fine.
Using a resizable memory buffer allows to allocate the right amount of
memory inside the backend, avoiding having to know the required buffer
size in advance.
The memory layout of the Mares Puck and Nemo devices is very similar,
which allows to share the parsing code between the backends.
The Mares Puck protocol allows for a more efficient implementation, by
reading only the data that we really need. But as an intermediate
solution, reusing the Nemo code is good enough.
